Closed dhimmel closed 7 years ago
I didn't track changes to:
modified: data/edges-10.sif
modified: data/edges-5k.sif.gz
modified: data/edges.sif.gz
modified: data/hetnet.json.bz2
modified: data/metagraph.json
modified: data/summary/degrees.xlsx
modified: viz/degrees.pdf
modified: data/edges-10.sif
modified: data/edges-5k.sif.gz
modified: data/edges.sif.gz
modified: data/hetnet.json.bz2
modified: data/metagraph.json
modified: data/summary/degrees.xlsx
modified: viz/degrees.pdf
It's a real bummer that these files changed. I don't think any of the changes are functional per se. Lot's of them I presume have to do with arbitrary dictionary ordering. This is a good reason to upgrade to Python 3.6 in the future, so dictionaries cease causing diff chaos.
@veleritas if you port the changes where I've mentioned you to your modified .ipynb
, I think you will be able to use hetio v0.2.3
without issue. This is my suggestion for solving https://github.com/dhimmel/hetio/issues/10. Let me know if you see any issues with this PR, otherwise I will merge.
@veleritas, I confirmed that this pull requests results in the exact same edges as Hetionet v1.0. I did not verify that edge properties are the same, but I think they should be.
Am going to merge. Let us know if porting the changes where I mention you resolves dhimmel/hetio#10 for you.
See https://github.com/dhimmel/hetio/issues/10.
hetio v0.2.3
raises an error if a duplicate edge is added as well as enables support for self loops. Previously, biderectional self loops would get filtered. This PR explicitly filters biderectional self loops to replicate the old behavior / hetnet.